AI Data Utilities
April 23, 2026 · View on GitHub
A curated list of open-source tools for data work — AI agents and chat-with-your-data interfaces, visualisation, cleaning and pipelines, dataframes, anomaly detection, geospatial utilities, reference datasets, and related resource lists.
Last updated: 2026-04-23. Updated periodically as new projects emerge.
Categories
- AI Agents
- Agent Skills
- MCP Servers
- Data Visualisation
- Data Cleaning
- DataFrames
- Anomaly Detection
- Geospatial
- Reference Data
- Resource Lists
AI Agents
AI agents, chat-with-your-data interfaces, and agentic data platforms.
MindsDB
Platform for building AI that can answer questions over large-scale federated data — connect hundreds of data sources and query them with natural language.
PandasAI
Chat with your data in natural language — conversational interface over Pandas dataframes and SQL databases with LLM-generated analysis and charts.
WrenAI
Open-source GenBI (Generative BI) agent — text-to-SQL, text-to-chart, and text-to-insights over your databases with a semantic-layer-aware approach.
DataLine
Self-hosted chat-with-your-database tool — natural-language SQL, charts, and data exploration with a focus on privacy and local-first use.
AI Data Science Team
A multi-agent "virtual data science team" — specialised agents for cleaning, feature engineering, modelling, and visualisation that collaborate on data-science workflows.
Vespa
Big-data serving engine for building AI-powered search, recommendation, and retrieval applications — hybrid vector + structured search at scale, widely used as the retrieval backbone for AI agents.
AI Analyst
Open-source AI data analyst — an agent that explores datasets, runs analyses, and produces narrative insights end-to-end.
Auto-Analyst
Agentic AI data-analytics system — multi-agent pipeline that automates data analysis, chart generation, and report writing over user-supplied datasets.
Agent Skills
Reusable skill packs that teach AI agents to work with specific data platforms or formats.
Markdown Viewer Skills
Collection of agent skills focused on Markdown viewing, parsing, and document workflows.
PlanetScale Database Skills
Official PlanetScale agent skills for working with their serverless MySQL database platform — schema, branching, and query workflows.
Microsoft Dataverse Skills
Microsoft's official agent skills for Dataverse — enabling AI agents to query, manipulate, and reason over Dataverse tables and records.
MCP Servers
Model Context Protocol servers that give AI agents structured access to data capabilities.
AntV MCP Server Chart
MCP server from the AntV team that lets LLMs generate charts and visualisations on demand — exposes a rich set of chart types as agent tools.
Data Visualisation
Charts, dashboards, data apps, and exploratory visual tools — AI-powered and classic.
Data Formulator
Microsoft Research tool for AI-powered data visualisation — iteratively create rich charts by combining a UI with natural-language transformation and re-shaping of data.
Erupt
Low-code admin / data-management framework — generate CRUD admin panels and dashboards over your data model with minimal configuration.
Perspective
Interactive analytics and visualisation component for streaming and large datasets — WebAssembly-powered pivot grids and charts that run in the browser, Jupyter, or server-side.
Bokeh
Python interactive visualisation library for modern web browsers — expressive graphics, dashboards, and data apps backed by a JavaScript rendering layer.
Orange
Visual programming environment for data mining, machine learning, and exploratory data analysis — drag-and-drop widgets for loading, transforming, modelling, and visualising data.
Constellation
Graph-focused data visualisation and analytic application — interactive exploration of large, complex network datasets.
Data Cleaning
Pipelines, transformation tools, and project scaffolds for preparing data.
Mage AI
Modern open-source data pipeline tool — build, run, and monitor ETL/ELT and transformation pipelines for cleaning and preparing data, with a UI-first workflow.
Cookiecutter Data Science
Standard project template for reproducible data-science work — opinionated folder structure, data-handling conventions, and tooling defaults.
Unity Catalog
Open-source universal catalog for data and AI — governance, lineage, and discovery across tables, files, functions, and ML models.
pyjanitor
Python clean-APIs extension for pandas — chainable, expressive data-cleaning verbs inspired by R's janitor package.
Enso
Hybrid visual + textual data-science language and notebook — build data pipelines by wiring nodes or writing code, with live data flow through each step.
DataFrames
Dataframe libraries and in-memory tabular engines.
pandas
The de-facto Python dataframe library — fast, flexible tabular data structures for data analysis and manipulation.
DataFrame
C++ statistical and financial dataframe library — Pandas-style heterogeneous columns with a focus on performance, with Python and Rust bindings.
Anomaly Detection
Tools and curated resources for detecting outliers and anomalous patterns in data.
Anomaly Detection Resources
Comprehensive curated list of anomaly / outlier detection papers, books, tutorials, datasets, and tools — a canonical starting point for the field.
Geospatial
GeoJSON tooling, maps, and geospatial data utilities.
geojson.io
Fast, simple, open-source editor for map data — draw, edit, and share GeoJSON in the browser.
python-geojson
Python bindings and utilities for GeoJSON — encode/decode, geometry classes, and feature manipulation.
Israel Municipalities Polygons
GeoJSON polygon boundaries for Israeli municipalities — useful for mapping, geofencing, and regional analysis in Israel.
Reference Data
Canonical lookup datasets — codes, lists, gazetteers.
iso3166 (PHP)
PHP library providing ISO 3166-1 country data — alpha-2, alpha-3, numeric codes, and names.
ISO-3166 Countries with Regional Codes
ISO 3166-1 country list enriched with UN M.49 regional / sub-regional / intermediate-region codes — delivered as CSV, JSON, XML, and YAML.
Israeli Cities and Streets List
Reference dataset of Israeli cities and their streets — useful for address autocomplete, form validation, and local-data applications.
Resource Lists
Other curated lists relevant to data work.
Awesome AI Data GitHub Repos
A broad curated index of AI and data-science GitHub repositories covering ML, data engineering, MLOps, and tooling.
Awesome Data Science
Long-running "awesome" list for data science — tutorials, books, podcasts, tools, datasets, and courses.
Awesome GeoJSON
Curated list of GeoJSON-related tools, libraries, converters, viewers, and datasets.
Contributing
Suggestions welcome — open an issue or PR with a link and a one-line description.
Disclaimer
Inclusion is not an endorsement. Verify licences, security, and fitness-for-purpose before using any listed project.